Fixes agents not spawning in last row and bumps version.

README.md

@@ -5,14 +5,17 @@
 <!-- TABLE OF CONTENTS -->
 ## Table of Contents
 
-* [About the Project](#about-the-project)
-  * [Built With](#built-with)
-* [Getting Started](#getting-started)
-  * [Installation](#installation)
-* [Usage](#usage)
-* [Citation](#citation)
-* [Contributing](#contributing)
-* [Contact](#contact)
+- [Level-Based Foraging Environment](#level-based-foraging-environment)
+  - [A multi-agent environment for Reinforcement Learning](#a-multi-agent-environment-for-reinforcement-learning)
+  - [Table of Contents](#table-of-contents)
+  - [About The Project](#about-the-project)
+    - [Built With](#built-with)
+  - [Getting Started](#getting-started)
+    - [Installation](#installation)
+  - [Usage](#usage)
+  - [Citation](#citation)
+  - [Contributing](#contributing)
+  - [Contact](#contact)
 
 
 
@@ -66,7 +69,7 @@ import lbforaging
 
 Then create an environment:
 ```python
-env = gym.make("Foraging-8x8-2p-1f-v0")
+env = gym.make("Foraging-8x8-2p-1f-v2")
 ```
 
 We offer a variety of environments using this template:
@@ -79,7 +82,7 @@ But you can register your own variation using (change parameters as needed):
 from gym.envs.registration register
 
 register(
-    id="Foraging-{0}x{0}-{1}p-{2}f{3}-v0".format(s, p, f, "-coop" if c else ""),
+    id="Foraging-{0}x{0}-{1}p-{2}f{3}-v2".format(s, p, f, "-coop" if c else ""),
     entry_point="lbforaging.foraging:ForagingEnv",
     kwargs={
         "players": p,

lbforaging.py

@@ -4,7 +4,7 @@
 import time
 import gym
 import numpy as np
-import lbforaging.foraging
+import lbforaging
 
 
 logger = logging.getLogger(__name__)
@@ -22,9 +22,8 @@ def _game_loop(env, render):
 
     while not done:
 
-        actions = []
-        for i, player in enumerate(env.players):
-            actions.append(env.action_space.sample())
+        actions = env.action_space.sample()
+
         nobs, nreward, ndone, _ = env.step(actions)
         if sum(nreward) > 0:
             print(nreward)
@@ -38,7 +37,7 @@ def _game_loop(env, render):
 
 
 def main(game_count=1, render=False):
-    env = gym.make("Foraging-8x8-2p-v0")
+    env = gym.make("Foraging-8x8-2p-2f-v2")
     obs = env.reset()
 
     for episode in range(game_count):

lbforaging/__init__.py

@@ -9,7 +9,7 @@
 
 for s, p, f, c, po in product(sizes, players, foods, coop, partial_obs):
     register(
-        id="Foraging{4}-{0}x{0}-{1}p-{2}f{3}-v1".format(s, p, f, "-coop" if c else "", "-2s" if po else ""),
+        id="Foraging{4}-{0}x{0}-{1}p-{2}f{3}-v2".format(s, p, f, "-coop" if c else "", "-2s" if po else ""),
         entry_point="lbforaging.foraging:ForagingEnv",
         kwargs={
             "players": p,

lbforaging/foraging/environment.py

@@ -258,8 +258,8 @@ def spawn_players(self, max_player_level):
             player.reward = 0
 
             while attempts < 1000:
-                row = self.np_random.randint(0, self.rows - 1)
-                col = self.np_random.randint(0, self.cols - 1)
+                row = self.np_random.randint(0, self.rows)
+                col = self.np_random.randint(0, self.cols)
                 if self._is_empty_location(row, col):
                     player.setup(
                         (row, col),

setup.py

@@ -2,7 +2,7 @@
 
 setup(
     name="lbforaging",
-    version="1.0.15",
+    version="1.0.16",
     description="Level Based Foraging Environment",
     author="Filippos Christianos",
     url="https://github.com/semitable/lb-foraging",

tests/test_env.py

@@ -10,7 +10,7 @@ def manhattan_distance(x,y):
 
 @pytest.fixture
 def simple2p1f():
-    env = gym.make("Foraging-8x8-2p-1f-v0")
+    env = gym.make("Foraging-8x8-2p-1f-v2")
     _ = env.reset()
     import time
 
@@ -29,7 +29,7 @@ def simple2p1f():
 
 @pytest.fixture
 def simple2p1f_sight1():
-    env = gym.make("Foraging-8x8-2p-1f-v0", sight=1)
+    env = gym.make("Foraging-8x8-2p-1f-v2", sight=1)
     _ = env.reset()
     import time
 
@@ -48,7 +48,7 @@ def simple2p1f_sight1():
 
 @pytest.fixture
 def simple2p1f_sight2():
-    env = gym.make("Foraging-8x8-2p-1f-v0", sight=2)
+    env = gym.make("Foraging-8x8-2p-1f-v2", sight=2)
     _ = env.reset()
     import time
 
@@ -67,18 +67,18 @@ def simple2p1f_sight2():
 
 
 def test_make():
-    env = gym.make("Foraging-8x8-2p-1f-v0")
-    env = gym.make("Foraging-5x5-2p-1f-v0")
-    env = gym.make("Foraging-8x8-3p-1f-v0")
-    env = gym.make("Foraging-8x8-3p-1f-coop-v0")
+    env = gym.make("Foraging-8x8-2p-1f-v2")
+    env = gym.make("Foraging-5x5-2p-1f-v2")
+    env = gym.make("Foraging-8x8-3p-1f-v2")
+    env = gym.make("Foraging-8x8-3p-1f-coop-v2")
 
 
 def test_spaces():
     pass
 
 
 def test_seed():
-    env = gym.make("Foraging-8x8-2p-2f-v0")
+    env = gym.make("Foraging-8x8-2p-2f-v2")
     for seed in range(10):
         obs1 = []
         obs2 = []
@@ -94,7 +94,7 @@ def test_seed():
 
 
 def test_food_spawning_0():
-    env = gym.make("Foraging-6x6-2p-2f-v0")
+    env = gym.make("Foraging-6x6-2p-2f-v2")
 
     for i in range(1000):
         _ = env.reset()
@@ -113,7 +113,7 @@ def test_food_spawning_0():
         assert foods[1][1] not in [0, 7]
 
 def test_food_spawning_1():
-    env = gym.make("Foraging-8x8-2p-3f-v0")
+    env = gym.make("Foraging-8x8-2p-3f-v2")
 
     for i in range(1000):
         _ = env.reset()